DJ Navarro

Work

About

Contact

Audible · Case Study

Designing against quiet value loss.

Audible has a million titles and a monthly credit, and no sense of how much time you actually have. This redesign targets three moments where that costs the listener something.

DSGN 310 · Team of 2 · My role: research lead, all interaction design and prototyping
The problem

What five listeners actually said

Five semi-structured interviews, 30 minutes each, with listeners ranging from a college student in Los Angeles to a retired travel planner in Bangkok.

5 of 5

Listened while doing something else.

Studying, commuting, driving eight hours a day, cleaning, falling asleep. Not one participant described sitting down to purely just listen. The app is designed for attention it never has.

4 of 5

Described friction with cost or credits.

In opposite directions; two forgot credits they had already paid, one bought three credits one month and none the next, and one for bought extra credits most months to cover an eight-hour driving day. Fixed monthly allotment frustrated them all.

3 of 5

Struggled to choose what to listen to.

With too many options, hard genre navigation, and recommendations that missed context and mood frustrates users. One participant described sometimes leaving the app entirely to look for title suggestions outside of Audible and losing traction to go back.

What we got wrong

Personal Companion vs Easier Decision-making

Our early predictions of planning a chatbot assumed users would want a full conversational companion, but interviews showed people wanted lightweight, low effort nudges. Five interviews and a survey later, we cut it. People didn't want the app to think for them, they wanted it to stop wasting what they'd already paid for and reduce the friction of finding and deciding on their best choice.

HMW

How might we?

Fifteen How Might We statements clustered into five problem spaces.

Credit value & generosity:

HMW let users redirect an unused credit to a friend so it feels like a gift, not a deadline?

Discovery & decision fatigue:

HMW give listeners a fast preview of a book's tone and narrator to decide in seconds.

Continuity & resumption:

HMW help listeners pick up exactly where they left off, even after a long break?

Context-aware personalization:

HMW give users curated, time-boxed collections ("fits your next 20 minutes")?

Reflection & growth:

HMW give listeners a low-effort way to jot notes or reflections after finishing?

Evidence

What the interviews actually support

Persona
Closest participant
What matches
Is this supported by interviewed Participants?

Casey Whitfield: the casual subscriber.

Aimee, 45, stay-at-home parent; Arwen, 21, student.

Aimee "often forgets about unused credits." Arwen "wished credits rolled over longer."

Supported (the forgetting and the expiry pressure both appear in the data).

Jessica Ramirez: the multitasking parent.

Aimee, 45, stay-at-home parent.

Listens while cleaning; wishes recommendations reflected her interests; short interrupted sessions.

Supported.

Mike Donovan: the commute listener.

Lance, 42, police officer.

Drives ~8 hours a day; Audible is what makes the commute survivable; buys extra credits.

Supported.

Lucas Okafor: the reflective listener.

No participant in this sample.

Note-taking, post-finish reflection, growth tracking.

Not supported: Lucas came from the hypothesis set, not the interviews.

Who I designed for

Four persona cards, 2 primary + 2 secondary.

The Casual Subscriber (primary)

Casey Whitfield

She binges a book in a week, then doesn't open the app for a month. Lets credits stack up, and then rushes a pick she doesn't want so the money isn't wasted.
Barrier: the monthly credit reads as a deadline, not a benefit.
Closest real analog: Aimee, 45, stay-at-home parent (often forgets about unused credits) and Arwen, 21, student (wished credits rolled over longer).

The Multitasking Parent (primary)

Jessica Ramirez

She listens in 15–30 minute windows around school drop-off, cleaning, and after bedtime. Doesn't browse; takes recommendations from parent groups and BookTok. 
Barrier: the catalog is larger than the time she has to evaluate it. 
Closest real analog: Aimee, 45 (listens while cleaning, wishes recommendations reflected her interests).

The Commute Listener (secondary)

Mike Donovan

He drives 30–45 minutes each way, five days a week. Loves long fantasy series but struggle to have time to search through book descriptions.
Barrier: browsing costs more effort than the listening is worth to him.
Closest real analog: Lance, 42, police officer (drives ~8 hours a day, buys extra credits).

The Reflective Listener (secondary)

Lucas Okafor

Nonfiction at 1.5–2x during workouts and study breaks. Gets a "what now" feeling after finishing. Writes notes on his phone, inconsistently.
Barrier: finishing a book produces nothing he can keep.
Closest real analog: none in this sample. Lucas came from our hypothesis set.

Architecture

Where new features go when the navigation is already full:

We added four things to an app that already had four tabs: a Credits & Gifting Hub, a mood picker and Quick Preview Reels under Discover, time-boxed collections, and Notes & Reflections under Library. Continue Listening was enhanced rather than added—it gained a return recap instead of a new home. Credits needed to be visible from anywhere, and the obvious move was a fifth tab. A fifth tab also degrades every other destination in the bar. I put the hub behind two entry points instead—a Credit Wallet widget on Home and a route through Profile—and accepted that discovery of the hub would be slower in exchange for not taxing the whole navigation. With the same logic, we kept Notes & Reflections inside Library rather than standing alone. The reflection is an artifact of a book you've finished; it belongs where the books are.

Design decisions

Design decisions in practice.

Expiry surfaced on Home, not buried in account settings.

Finding: Aimee (45) "often forgets about unused credits." Arwen (21) "wished credits rolled over longer." The credit was invisible until it was urgent.
Response: a persistent Home strip reading "1 credit · expires in 5 days" with the expiry date, not just the balance.
Cost: it puts a countdown on the home screen of an app people open to relax. I chose a single quiet line over a badge or a push notification, because the goal was awareness, not pressure as the pressure is the thing we're trying to remove.

Gifting as an alternative to forfeiting.

Finding: the credit model is mismatched to consumption in both directions. Lance (42, ~8 hrs driving/day) buys extra credits most months, and Aimee forgets hers. The same fixed monthly allotment fails a heavy user and a light user in opposite ways.
Response: Gift a credit—pick a contact, add a note, send now or schedule—sitting one tap from the balance.
Cost: gifting is the one feature here I can't source to a participant. It came from a survey question we wrote to test it. I built it because it's the only lever that converts guilt into generosity, and I'd want it tested before shipping.

Time-based discovery: "How much time do you have?"

Finding: all five participants listened while doing something else — studying, driving eight hours, cleaning, falling asleep. Not one described sitting down to listen. The real constraint on a pick isn't genre, it's the length of the gap.
Response: Discover leads with a time filter (under 20 min / 20–45 / 1–3 hrs / 3+) and a "Fits your next 20 minutes" collection, above the catalog.
Cost: organizing by duration demotes genre browsing, which is how Emily said she looks for books. I traded her use case for the more common one and would test whether that's the right call.

Quick Preview Reels: a unanimous desire.

Finding: narration was the only thing all five participants agreed on. Every one of them brought it up unprompted as the best part of the product. Meanwhile three of five described stalling at the point of choosing—too many options, hard genre navigation, or recommendations that missed context. The app was hiding its strongest selling point by only using description text and star ratings.
Response: Quick Preview Reels, a 15–30 second clip of the book’s narrator, surfaced as a row under Discover and attached to every title in a mood or time-boxed collection. Tapping one opens a full-screen player with the narrator audio, the runtime, and a single Start listening action against the credit.
Cost: reels don't scale to a million titles. Whatever subset gets one is effectively promoted, which turns a discovery aid into a merchandising surface—and the titles most likely to get produced reels are the ones that need the help least. I also demoted written description and reviews to make room, which is how Emily said she evaluates a book. I'd want to know whether reels raise completion rate or just raise the rate at which people start things they abandon.

What I explored

[Heading goes here]

[Paragraph goes here]

How I’d test it

[Heading goes here]

[Paragraph goes here]

[Test plan line]

Limitations

[Heading goes here]

[Paragraph goes here]